[1] For the determination of the stress at the air-sea interface and the near-surface wind speed, numerical weather prediction (NWP) models commonly use a drag coefficient. Generally, the Charnock relation is used, which gives an increase in the drag coefficient for increasing nearsurface wind speeds. According to observations, however, the magnitude of the drag coefficient levels off at a wind speed of approximately 30 m s −1 , and decreases with a further increase of the wind speed. Consequently, the surface drag is overestimated in NWP models for hurricane wind speeds and the intensity of hurricane winds is underestimated in forecasts. In this study, a parameterization that gives a decrease in the surface drag is tested in an NWP model. Two hurricanes in the Caribbean are modeled: Ivan (2004) and Katrina (2005). The results show that this drag parameterization leads to much stronger hurricanes in forecasts, and are in good agreement with observations. Citation: Zweers, N. C., V. K. Makin, J. W. de Vries, and G. Burgers (2010), A sea drag relation for hurricane wind speeds, Geophys.

Abstract.To simulate winds and water levels, numerical weather prediction (NWP) and storm surge models generally use the traditional bulk relation for wind stress, which is characterized by a wind drag coefficient. A still commonly used drag coefficient in those models, some of them were developed in the past, is based on a relation, according to which the magnitude of the coefficient is either constant or increases monotonically with increasing surface wind speed (Bender, 2007;Kim et al., 2008;Kohno and Higaki, 2006). The NWP and surge models are often tuned independently from each other in order to obtain good results. Observations have indicated that the magnitude of the drag coefficient levels off at a wind speed of about 30 m s −1 , and then decreases with further increase of the wind speed. Above a wind speed of approximately 30 m s −1 , the stress above the air-sea interface starts to saturate. To represent the reducing and levelling off of the drag coefficient, the original Charnock drag formulation has been extended with a correction term.In line with the above, the Delft3D storm surge model is tested using both Charnock's and improved Makin's wind drag parameterization to evaluate the improvements on the storm surge model results, with and without inclusion of the wave effects. The effect of waves on storm surge is included by simultaneously simulating waves with the SWAN model on identical model grids in a coupled mode. However, the results presented here will focus on the storm surge results that include the wave effects.The runs were carried out in the Gulf of Mexico for Katrina and Ivan hurricane events. The storm surge model was initially forced with H * wind data (Powell et al., 2010) to test the effect of the Makin's wind drag parameterization on the storm surge model separately. The computed wind, water levels and waves are subsequently compared with observation data. Based on the good results obtained, we conclude that, for a good reproduction of the storm surges under hurricane conditions, Makin's new drag parameterization is favourable above the traditional Charnock relation. Furthermore, we are encouraged by these results to continue the studies and establish the effect of improved Makin's wind drag parameterization in the wave model.The results from this study will be used to evaluate the relevance of extending the present towards implementation of a similar wind drag parameterization in the SWAN wave model, in line with our aim to apply a consistent wind drag formulation throughout the entire storm surge modelling approach.

No abstract

The impact of new parametrizations for drag and air-sea enthalpy exchange on modelling the intensity of tropical cyclones with a numerical weather prediction (NWP) model is examined. These parametrizations follow from a model for the marine atmospheric boundary layer for high wind-speed conditions in the presence of spray droplets that originate from breaking wave crests. This model accounts for the direct impact of these droplets on the air-sea momentum flux through action of a spray force, which originates from the interaction of the 'rain' of spray droplets with the vertical wind shear and is expressed in terms of the spray generation function (SGF). The SGF is cubic in the wind speed up to 50 m s −1 beyond which its value increases less strongly. The drag coefficient (C D ) decreases from a wind speed of approximately 30 m s −1 , in agreement with the available measurements in these conditions. The enthalpy exchange coefficient (C k ) increases with increasing wind speed and slowly decreases beyond a wind speed of about 40 m s −1 due to the strong decrease in C D . The value for C k /C D is in agreement with observational data for wind speeds up to 30 m s −1 ; for higher wind speeds the value is in the range 1.2-1.5 in agreement with a well-established theory. The parametrization is tested in an NWP model. The tropical cyclones Ivan (2004) and Katrina (2005) in the Gulf of Mexico are simulated. To the sea surface temperatures (SSTs) from the European Centre archive that were prescribed to the NWP model, a parametrized cooling (based on estimations from theoretical studies and measurements) was applied during the V. N. Kudryavtsev Russian State Hydrometeorological University (RSHU), Saint Petersburg, Russia 123 N. C. Zweers et al. model forecasts, as the NWP model does not resolve locally rather strong induced reductions in SSTs. The simulations show that realistic tropical cyclone wind speeds and central pressure can be obtained with the proposed drag and enthalpy parametrizations. The results indicate that the value for C k /C D at very high wind speeds is in the correct range. Moreover, the results motivate the application of the parametrizations in atmosphere-ocean coupled models.